Machine Learning

Pioneer machine learning researcher Arthur Samuel defined machine learning as: “the field of study that gives computers the ability to learn without being explicitly programmed”.

Tom M. Mitchell provided a widely quoted, more formal definition: “A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P, if its performance at tasks in T, as measured by P, improves with experience E”.

Machine Learning is a subfield of computer science and engineering which evolved from the computer learning theory and the study of pattern recognition in AI (Artificial Intelligence). Machine learning studies computer algorithm for learning to do some stuff. It explores the construction and study of algorithms that can learn from and also make a prediction on data. The process of learning data is being done is always depend on some factors, observations or data, such as direct experience, examples or instructions as well. So machine learning is all about learning to do better in future based on what was experienced in previously. In this article, we’ll explain some examples of machine learning technology.

 

Samual wrote a checkers program back in 1959, which learned to play chekers slowly and become better player than Samual.

Some examples of Machine Learning:

1. Face Detection: used to recognize faces in images or indicate if a face is present in an image or not.
2. Topic Spotting: used to categorize articles as to whether they are about sports, entertainments, politics or science etc.
3. Spam filtering: recognize if an email message is a spam or a non-spam.
4. Optical character recognition: used to categorize images of handwritten characters from the collection of letters.
5. Medical diagnosis: It’ll also diagnose a patient as a sufferer or non-sufferer of any disease.
6. Weather prediction: It’ll predict the weather for today and upcoming days as well.
7. Fraud detection: used to identify if a transaction such as credit card transaction is fraudulent in nature.
8. Suggestion from shopping site

There are some most popular inventions that uses the concept of machine learning algorithms such as:

Google Car uses machine learning: The developers of Google car use the machine learning algorithms to create a prototype model of objects and person on the road. Once it knows that the cars are in space, it can initiate the work of developing and modeling the behavior of other dynamic objects such as bicycles, cars, person, and pedestrians. Every single mile of driving is logged and the data stored into computers that classify how different objects behave in a variety of situations.

IBM Watson: IBM Watson is another example that uses the machine learning but for limited behavior. It’s regression feature doesn’t work on large datasets. It ingestion of more than 600,000 medical evidences, more than 2 million pages from medical practices and search through more than 1.5 million patients records for further information gives it a huge amount of knowledge, no human doctor can ever match. IBM Watson is already capable of collecting and storing far more medical information than normal human doctors and its decisions are all evidence-based and free of overconfidence and cognitive biases.

Robotic Process Automation: Robotics process automation (RPA) is the output of the technology which allows employees in a company to configure a “robot” or computer software to understand and interpret existing software application for manipulating data, processing any transaction, communicating with other digital system and the triggering wide range of responses. These robots are learning to respond automatically for an event using their past experience. They help to reduce time, create user-friendly environment and increase the growth.